simple script to read a TSV file and creaate author pages for Hugo/Wowchemy/Academic theme

import_authors_for_wowchemy.py

#!/usr/bin/env python
#### import_authors.py
# python 3.7+
# create author pages for a Hugo/Wowchemy site ( https://wowchemy.com/ )
# from a  TSV file (e.g. download from google sheets) with very specific headers
# import_authors.py <tsv file> <../path/to/site>
# requires a TSV with the following field header row: 
# Name	ProjectRole	Photo	InstitutioneRole/Title	URL	Org1Name	Org1URL	Org2Name	Org2URL	one_line_bio	interests	PhD	PhDInstitution	PhDYear	MS	MSInstitution	MSYear	BS/BA	BS/BAInstitution	BS/BAYear	emailaddress	twitterid	githuburl	googlescholarurl	paragraph_bio

# clunky but works for now! 

import csv, os.path, argparse, warnings
# jinja2 is only external lib used 
from jinja2 import Template, Environment, FileSystemLoader, select_autoescape


def tsv_to_authors(people_file, site_folder, author_path="content/authors"):

    """read in tsv file, for each row create an author entry (index.md) 
    for the Hugo template system from 'wowchemmy'
    formerly academia.  
    
    1, Creates author folders by using only alphanumeric chars from author names
    2. creates index.md file in each folder OVERWRITES FILES WITHOUT WARNING
    does not delete any existing folders
    does not copy photos over
    requires a template file
    requires column names to match template
    """
    author_path = os.path.join(site_folder, "content/authors")
    #env = Environment(
    #    loader=FileSystemLoader(template_path),
    #    autoescape=select_autoescape(['html', 'xml']),
    #)
    #template = env.get_template(template_file)

    template = Template(author_template)
    with open(people_file, "r") as f:
        reader = csv.DictReader(f, delimiter="\t") # assumes TSV
        for p in reader:
            print(p['Name'])
            author_folder = "".join([c for c in p['Name'] if c.isalpha()]).rstrip()        
            author_file = os.path.join(author_path, author_folder, '_index.md')
            author_md = template.render(p=p)
            # print(author_md)
            os.makedirs(os.path.join(author_path, author_folder),exist_ok=True)
            with open(author_file, 'w') as the_file:
                the_file.write(author_md)

author_template = """
---
# Display name
title: {{ p['Name'] }}

# Is this the primary user of the site?
superuser: false

# Role/position
role:  {{ p['InstitutioneRoleTitle'] }}

# Organizations/Affiliations
organizations:
- name: {{ p['Org1Name'] }}
  url: "{{ p['Org1URL'] }}"
{% if p['Org2Name']|length -%}
- name: {{ p['Org2Name'] }}
  url: "{{ p['Org2URL'] }}" 
{%- endif %}
# Short bio (displayed in user profile at end of posts)
bio: {{p['one_line_bio']}}

{% if p['interests']|length %}
interests:
{% for interest in p['interests'].split(',') -%}
- {{interest.strip()}} 
{% endfor %}
{%- endif %}

education:
  courses:
  - course: {{ p['PhD'] }}
    institution: {{ p['PhDInstitution'] }}
    year: {{p['PhDYear']}}
{%- if p['MS']|length %}
  - course: {{p['MS']}}
    institution: {{p['MSInstitution']}}
    year: {{p['MSYear']}}
{%- endif %}
{%- if p['BS/BA']|length %}
  - course: {{p['BS/BA']}}
    institution: {{p['BS/BAInstitution']}}
    year: {{p['BS/BAYear']}}
{%- endif %}

# Social/Academic Networking
# For available icons, see: https://sourcethemes.com/academic/docs/page-builder/#icons
#   For an email link, use "fas" icon pack, "envelope" icon, and a link in the
#   form "mailto:[email protected]" or "#contact" for contact widget.
social:
- icon: envelope
  icon_pack: fas
  link: "mailto:{{ p['emailaddress'] }}"
{% if p['twitterid']|length -%}
- icon: twitter
  icon_pack: fab
  link: "https://twitter.com/{{p['twitterid']}}"
{%- endif %}
{%- if p['googlescholarurl']|length %}
- icon: google-scholar
  icon_pack: ai
  link: "{{ p['googlescholarurl']}}"
{%- endif -%}
{%- if p['githuburl']|length %}
- icon: github
  icon_pack: fab
  link: "{{p['githuburl']}}"
{%- endif -%}
# Link to a PDF of your resume/CV from the About widget.
# To enable, copy your resume/CV to `static/files/cv.pdf` and uncomment the lines below.
# - icon: cv
#   icon_pack: ai
#   link: files/cv.pdf

# Enter email to display Gravatar (if Gravatar enabled in Config)
# email: "{{ p['emailaddress'] }}"

# Highlight the author in author lists? (true/false)
highlight_name: false

# Organizational groups that you belong to (for People widget)
#   Set this to `[]` or comment out if you are not using People widget.
user_groups:
- Group Members
{%- if p['Role'] -%}
- p['Role']
{%- endif %}
---

{{p['paragraph_bio']}}

"""


if __name__ == "__main__":
    
    parser = argparse.ArgumentParser()
    parser.add_argument('people_file', type=str,help="TSV formatted file")
    parser.add_argument('site_path',
                        type=str,
                        help='root folder of the Academic/Hugo website')

    args = parser.parse_args()

    # site_folder = "../../srm-ecology-website"
    # people_file="srm-ecology-people.tsv"
    # template_file="authortemplate.md"
    print(f"importing tsv {args.people_file} to {args.site_path}")
    
    tsv_to_authors(people_file=args.people_file, site_folder=args.site_path)